On the Improvement of the Reasoning Cycle in Case-Based Reasoning

Case-based reasoning (CBR) is undoubtedly by far one of the most intuitive artificial intelligence problem-solving approaches. It is inherent to the reuse of existing experience that includes solutions to problems or mechanisms to derive these solutions. Unfortunately, existing CBR systems are lacking in generality as the adaptation process is usually driven by the application domain and heavily based on the domain expert's knowledge. Moreover, existing CBR systems perform poorly as they deal with each step of the CBR methodology separately and independently from the other steps. This work is an effort to lay a primary foundation for a generic case-based reasoning framework that relies on a domain-independent approach. Each step of the reasoning process is conceived to fulfil not only specific requirements of that stage but also to sustain other stages' processes. The experimental results of the application of our approach in a sober consumption energy system in buildings show the effectiveness of our approach.